Clare Island Survey — Phanerogamia. 10 75 



apparatus, where their arrival was observed on a sheet of black or white 

 paper strongly illuminated. The behaviour of the seeds in their downward 

 course could be watched by an observer stationed at the top. Freshly 

 gathered seed was used wherever possible ; in only a few cases was seed a 

 season or more old employed. Comparison made as to the rate of fall of old 

 and new seed of a number of species showed no appreciable difference 

 between the two. This result corresponds with that obtained in his flotation 

 experiments by G-uppy, who found very little difference in buoyancy between 

 fresh and dried seeds. Twenty to forty seeds, taken in almost all cases at 

 different dates from several different plants, or from different portions of the 

 same plant, were tested in the case of each species. Care was taken to see 

 that the pappus or wing (when present) was complete and uninjured, and 

 that the seed was mature ; in a few doubtful cases the seeds were afterwards 

 germinated. To the naked eye, a difference in the size and shape of the seed 

 or of the pappus or wing was often obvious, producing even in seeds selected 

 for their perfect pappus a variation in the velocity of fall sometimes 

 amounting to as much as 50 per cent. ; in the majority of cases the results 

 were uniform as regards seeds of any one inflorescence ; but the degree of 

 expansion of the pappus was found to vary greatly even in seeds quite 

 mature and in the course of being shed naturally, and this sometimes caused 

 a variation of as much as 100 per cent, in the velocity of fall; in such cases 

 the seeds tested were those in which the pappus was most expanded. It was 

 found that warming the seeds before a fire immediately before the experiment 

 tended to expand the pappus and increase the efficiency. This would happen 

 in nature in hot sunlight. 



Where the plume is stiff as in most of the pappus-bearing Compositae, 

 there is but little variation in the rate of fall ; but where it is lax, as in the 

 Willow-herbs, Willows, and Cotton-grasses, the variation is much greater, 

 and a number of different gatherings ought to be tested — more in some cases 

 -than I had at my disposal — before one can feel assured that the maximum 

 efficiency has been found. Probably, in order to obtain the practical 

 maximum efficiency of seeds matured and liberated under the most favour- 

 able conditions, we might safely add 5 per cent, to the figures given below 

 for stiff-pappus seeds, and 10 per cent, for the lax -pappus seeds. 



In some minute seeds also, a large variation in the rate of fall was found. 

 In some of the Orchidaceae, for instance, this amounted to as much as 

 100 per cent., depending apparently on variation in the disposition of the 

 loose netted testa. The difference in the proportion of the figures below 

 for a 12-foot and a 40-foot fall, may in some cases be due to the fact that 

 different batches of seed were experimented on in the two cases. 



K2 



